Deep Learning Appliance

DEEP Gadget provides a silent, reliable, and powerful deep learning platform powered by up to seven water-cooled GPUs and ready-to-use software stack.

Configure & Order

Features

Best-In-Class Low Noise Design

DEEP Gadget works quietly even at full load because water cooling significantly reduces fan noise. It can even be placed in your office.

Incredible Computing Power of 7 GPUs

7 GPUs in a single DEEP Gadget system provides up to 100 TFLOPS of single-precision floating-point performance, and 700 TFLOPS of deep learning performance by Volta Tensor Core. A closed-circuit water cooling system keeps high-density GPUs at a low temperature.

Ready-To-Use Deep Learning Software Stack

DEEP Gadget is delivered with pre-installed software stacks including an operating system and deep learning frameworks, such as Caffe and TensorFlow.

Pricing

DEEP Gadget uses water cooling to ensure enterprise-level lifetime and reliability in cost-effective gaming GPU.

1.68x powerful than a $49,000 server w/ 4x V100
$29,070
7x NVIDIA RTX 2080
20 Intel Xeon Scalable CPU cores
128 GB main memory
250 GB primary storage
12 TB secondary storage
$34,845
7x NVIDIA RTX 2080 Ti
20 Intel Xeon Scalable CPU cores
128 GB main memory
250 GB primary storage
12 TB secondary storage
$53,822
7x NVIDIA TITAN V
20 Intel Xeon Scalable CPU cores
128 GB main memory
250 GB primary storage
12 TB secondary storage

All prices are subject to change without any prior notice. Tax and shipping cost are not included.

Specification

Items Description
Form Factor Standalone or 4U Rackmount
# of CPU Sockets 1 or 2
CPU

One of the following:

  • Intel Xeon Scalable Gold 5120, 5122, 6128, 6130, 6134, 6138, 6140, 6142, 6148, or 6152
  • Intel Xeon Scalable Silver 4108, 4110, 4114, or 4116
  • Intel Xeon Scalable Bronze 3104 or 3106
  • Intel Xeon E5-2603 v4, 2609 v4, 2620 v4, 2630 v4, 2640 v4, 2650 v4, 2660 v4, 2680 v4, 2683 v4, or 2690 v4
Maximum # of GPUs 7
GPU

One of the following:

  • NVIDIA GeForce RTX 2080
  • NVIDIA GeForce RTX 2080 Ti
  • NVIDIA TITAN V
  • NVIDIA Tesla P100 PCIe 16 GB
  • NVIDIA Tesla V100 for PCIe 16 GB or 32 GB
PCIe
  • 3x PCIe 3.0 x16 and 4x PCIe 3.0 x8 if two Intel Xeon Scalable CPUs are installed
  • 2x PCIe 3.0 x16 and 5x PCIe 3.0 x8 if two Intel Xeon E5-2600 v4 CPUs are installed
  • 1x PCIe 3.0 x16 and 6x PCIe 3.0 x8 if a single CPU is installed
Main Memory DD4 2400 MHz ECC Registered, 64 GB, 128 GB, 192 GB, 256 GB, 384 GB, 512 GB, or 768 GB
(192 GB or more main memory is applicable only if two Intel Xeon Scalable CPUs are installed)
Primary Storage 250 GB, 500 GB, 1 TB, or 2 TB M.2 NVMe SSD
Secondary Storage
(Optional)

One of the following for a standalone model:

  • 8 TB SATA3 HDD
  • 10 TB SATA3 HDD
  • 12 TB SATA3 HDD
  • 14 TB SATA3 HDD
  • 32 TB RAID 6 HDD storage (6x 8 TB SATA3 HDD, Software RAID 6 4+2)
  • 40 TB RAID 6 HDD storage (6x 10 TB SATA3 HDD, Software RAID 6 4+2)
  • 48 TB RAID 6 HDD storage (6x 12 TB SATA3 HDD, Software RAID 6 4+2)
  • 56 TB RAID 6 HDD storage (6x 14 TB SATA3 HDD, Software RAID 6 4+2)
  • 4 TB SATA3 SSD
  • 16 TB RAID 6 SSD storage (6x 4 TB SATA3 SSD, Software RAID 6 4+2)

One of the following for a rackmount model:

  • 4.8 TB RAID 6 HDD storage (6x 1.2 TB SATA3 HDD, Software RAID 6 4+2)
  • 7.2 TB RAID 6 HDD storage (6x 1.8 TB SATA3 HDD, Software RAID 6 4+2)
  • 9.6 TB RAID 6 HDD storage (6x 2.4 TB SATA3 HDD, Software RAID 6 4+2)
  • 14.4 TB RAID 6 HDD storage (8x 2.4 TB SATA3 HDD, Software RAID 6 6+2)
  • 4 TB SATA3 SSD
  • 16 TB RAID 6 SSD storage (6x 4 TB SATA3 SSD, Software RAID 6 4+2)
  • 24 TB RAID 6 SSD storage (8x 4 TB SATA3 SSD, Software RAID 6 6+2)
Motherboard ASUS WS X299 SAGE, WS C621E SAGE, X99-E WS, or Z10PE-D8 WS
Power Supply 2x 1300 Watts (80 PLUS Platinum)
Operating System Ubuntu 18.04 LTS
Libraries CUDA Toolkit, cuBLAS, cuDNN, NCCL, NCCL 2
Deep Learning Frameworks Caffe, TensorFlow, Caffe2, Torch, PyTorch, Theano, MXNet, Cognitive Toolkit

Our Valued Customers

You can customize a DEEP Gadget system and make an order online.

Configure & Order

Room 308, Building 138, Seoul National University, Seoul 08826, Korea
sales@manycoresoft.co.kr

Copyright © 2018 ManyCoreSoft Co., Ltd.